[
https://issues.apache.org/jira/browse/HDFS-7611?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14295956#comment-14295956
]
Konstantin Shvachko commented on HDFS-7611:
-------------------------------------------
Jing, I looked closer. That is added more logging. I think there is something
going with delete itself, not snapshot delete.
On the positive side I was running this with pre-HDFS-7676 version of
TestFileTruncate and it never failed.
So I think this particular problem is fixed. Let's file another jira for
TestOpenFilesWithSnapshot, if there isn't one already.
+1 for your patch.
> deleteSnapshot and delete of a file can leave orphaned blocks in the
> blocksMap on NameNode restart.
> ---------------------------------------------------------------------------------------------------
>
> Key: HDFS-7611
> URL: https://issues.apache.org/jira/browse/HDFS-7611
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: namenode
> Affects Versions: 2.6.0
> Reporter: Konstantin Shvachko
> Assignee: Byron Wong
> Priority: Critical
> Attachments: HDFS-7611.000.patch, HDFS-7611.001.patch,
> HDFS-7611.002.patch, blocksNotDeletedTest.patch, testTruncateEditLogLoad.log
>
>
> If quotas are enabled a combination of operations *deleteSnapshot* and
> *delete* of a file can leave orphaned blocks in the blocksMap on NameNode
> restart. They are counted as missing on the NameNode, and can prevent
> NameNode from coming out of safeMode and could cause memory leak during
> startup.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
